Reading the repo
Opening the box…
Reading the repo
Opening the box…
How to install on Linux
Open-source registry and CLI for Claude Code components: agents, commands, skills, hooks, MCPs, settings and mods
Saved setup from 3 h ago
Will it work on Linux?
Also: Docs ↗
Heads-upBuilding it yourself? It also needs 20 settings in .env (9 secret).
CONTRIBUTING.mdWe welcome contributions! Help us make Claude Code even better for everyone.
📋 Before contributing, please read our Code of Conduct to ensure a respectful and inclusive environment for all community members.
The easiest way to contribute is by adding individual components: agents, commands, skills, MCPs, settings, hooks, mods or loops.
scripts/generate_components_json.py and do not commit docs/components.json or anything under dashboard/public/. Those files are regenerated automatically after merge, and a PR that touches them fails the generated-files-guard check.claude plugin validate), NVIDIA SkillSpector security scanning and AI code review. A maintainer or the daily triage then labels it triage:accept, triage:changes (see the comment for what to fix), triage:human (needs a maintainer's manual review: hooks with scripts, mods, anything that runs code or sends data off the machine) or triage:reject (with the reason).Agents are AI specialists for specific domains (security, performance, frameworks, etc.).
Create Agent File
# Navigate to appropriate category
cd cli-tool/components/agents/[category]/
# Create your agent file
touch your-agent-name.md
Agent File Structure
# Agent Name
Agent description and purpose.
## Expertise
- Specific domain knowledge
- Key capabilities
- Use cases
## Instructions
Detailed instructions for Claude on how to act as this agent.
## Examples
Practical examples of agent usage.
Available Categories
development-team/ - Full-stack developers, architectsdomain-experts/ - Security, performance, accessibility specialistscreative-team/ - Content creators, designersbusiness-team/ - Product managers, analystsdevelopment-tools/ - Tool specialists, DevOps expertsCreating New Categories If your agent doesn't fit existing categories, create a new one:
# Create new category folder
cd cli-tool/components/agents/
mkdir your-new-category
# Add your agent file to the new category
cd your-new-category/
touch your-agent-name.md
Commands are custom slash commands that extend Claude Code functionality.
Create Command File
cd cli-tool/components/commands/[category]/
touch your-command-name.md
Command File Structure
# /command-name
Brief command description.
## Purpose
What this command accomplishes.
## Usage
How to use the command with examples.
## Implementation
Technical details of what the command does.
Command Categories
code-generation/ - Generate code, tests, documentationanalysis/ - Code analysis, optimization, debuggingproject-management/ - File operations, project structuretesting/ - Test generation, validation, coveragedeployment/ - Build, deploy, CI/CD operationsCreating New Categories If your command doesn't fit existing categories, create a new one:
# Create new category folder
cd cli-tool/components/commands/
mkdir your-new-category
# Add your command file to the new category
cd your-new-category/
touch your-command-name.md
MCPs provide external service integrations for Claude Code.
Create MCP File
cd cli-tool/components/mcps/[category]/
touch your-service-mcp.json
MCP File Structure
{
"mcpServers": {
"service-name": {
"command": "npx",
"args": ["-y", "@your-org/mcp-server"],
"env": {
"API_KEY": "<YOUR_API_KEY>",
"BASE_URL": "https://api.service.com"
}
}
}
}
MCP Categories
audio/ - Audio processing, text-to-speech, transcription servicesintegration/ - GitHub, GitLab, Jiradatabase/ - PostgreSQL, MySQL, MongoDBcloud/ - AWS, Azure, GCP servicesdevtools/ - Build tools, testing frameworksai-services/ - OpenAI, Anthropic, other AI APIsCreating New Categories If your MCP doesn't fit existing categories, create a new one:
# Create new category folder
cd cli-tool/components/mcps/
mkdir your-new-category
# Add your MCP file to the new category
cd your-new-category/
touch your-service-mcp.json
Settings configure Claude Code behavior and performance.
Create Settings File
cd cli-tool/components/settings/[category]/
touch your-setting-name.json
Settings File Structure
{
"setting-category": {
"parameter": "value",
"description": "What this setting controls"
}
}
Settings Categories
performance/ - Memory, timeout, cache settingsui/ - Interface customization, themesmcp/ - MCP server configurationssecurity/ - Access control, permissionsCreating New Categories If your setting doesn't fit existing categories, create a new one:
# Create new category folder
cd cli-tool/components/settings/
mkdir your-new-category
# Add your setting file to the new category
cd your-new-category/
touch your-setting-name.json
Hooks provide automation triggers for different development events.
Create Hook File
cd cli-tool/components/hooks/[category]/
touch your-hook-name.json
Hook File Structure
{
"hooks": {
"hook-name": {
"event": "trigger-event",
"command": "action-to-perform",
"description": "What this hook does"
}
}
}
Hook Categories
git/ - Pre-commit, post-commit, pre-pushdevelopment/ - File changes, build eventstesting/ - Test execution, coverage checksCreating New Categories If your hook doesn't fit existing categories, create a new one:
# Create new category folder
cd cli-tool/components/hooks/
mkdir your-new-category
# Add your hook file to the new category
cd your-new-category/
touch your-hook-name.json
Skills are folders under cli-tool/components/skills/{category}/{name}/ with a SKILL.md (YAML frontmatter with name and description) and any supporting files.
name in the frontmatter must match the folder name (kebab-case).description says what the skill does and when Claude should use it..claude-plugin/ or .claude/skills/: those paths are not part of the catalog.source.json pinned to a commit instead of copying it (see "External skills" in CLAUDE.md).Test it with npx claude-code-templates@latest --skill {category}/{name} --dry-run.
Mods (Claude Mods) are plugin directories under cli-tool/components/mods/{category}/{name}/: .claude-plugin/plugin.json, hooks/hooks.json, the hooks modules under hooks/, a README.md, and optionally commands/, types/ and tests/. Before opening the PR:
cd cli-tool/components/mods
npx -y -p typescript@5 tsc -p tsconfig.json # must typecheck against types/claude-code.d.ts
claude plugin validate {category}/{name}
claude plugin test {category}/{name} # if the mod ships tests
The README must state the command, the options and the minimum Claude Code version. Mods always get a maintainer's manual review because they run inside Claude Code.
Loops are autonomous workflows (goal + interval + stop condition) under cli-tool/components/loops/{category}/{name}.md. They may reference other catalog components, which are installed with the loop, so every referenced component must already exist in the catalog.
Templates are complete project configurations that include CLAUDE.md, .claude/* files, and .mcp.json.
Create Template Directory
cd cli-tool/templates/
mkdir your-template-name
cd your-template-name
Template Structure
your-template-name/
├── CLAUDE.md # Main configuration
├── .claude/
│ ├── settings.json # Automation hooks
│ └── commands/ # Template-specific commands
├── .mcp.json # MCP server configuration
└── README.md # Template documentation
CLAUDE.md Guidelines
Template Categories
For advanced contributors who want to improve the CLI tools like analytics, health check, and chat monitoring.
# Clone the repository
git clone https://github.com/davila7/claude-code-templates.git
cd claude-code-templates
# Navigate to the CLI tool directory
cd cli-tool
# Install dependencies
npm install
# Link for local testing
npm link
# Run test suite
npm test
The analytics dashboard provides real-time monitoring of Claude Code sessions.
# Start analytics dashboard
npm run analytics:start
# Clear cache during development
curl -X POST http://localhost:3333/api/cache/clear -H "Content-Type: application/json" -d '{"type":"all"}'
# Refresh data
curl http://localhost:3333/api/refresh
# Restart server completely
pkill -f analytics && sleep 3 && npm run analytics:start
src/analytics/
├── core/ # Business logic
│ ├── StateCalculator.js # Conversation state detection
│ ├── ProcessDetector.js # Running process detection
│ ├── ConversationAnalyzer.js # Message parsing
│ └── FileWatcher.js # Real-time file monitoring
├── data/ # Data management
│ └── DataCache.js # Multi-level caching
├── notifications/ # Real-time communication
│ ├── WebSocketServer.js # Server-side WebSocket
│ └── NotificationManager.js # Event-driven notifications
└── utils/ # Utilities
└── PerformanceMonitor.js # System health monitoring
Problem: Changes don't appear in dashboard
# Solution: Clear cache and refresh
curl -X POST http://localhost:3333/api/cache/clear -H "Content-Type: application/json" -d '{"type":"conversations"}'
curl http://localhost:3333/api/refresh
Problem: WebSocket not updating
# Solution: Hard refresh browser (Ctrl+F5 or Cmd+Shift+R)
Mobile-optimized interface for viewing Claude conversations in real-time.
src/chats/
├── components/ # UI components
├── services/ # API communication
├── websocket/ # Real-time updates
└── styles/ # Mobile-first CSS
# Start chat monitor
npm run chats:start
# Start with tunnel (requires cloudflared)
npm run chats:start -- --tunnel
# Test mobile interface
npm run chats:test
Comprehensive diagnostics tool for Claude Code installations.
# Run health check
npm run health-check
# Add new health check
# 1. Create check in src/health-checks/
# 2. Add to health check registry
# 3. Test with various scenarios
# Test component installation
npx claude-code-templates@latest --agent your-agent --dry-run
npx claude-code-templates@latest --command your-command --dry-run
npx claude-code-templates@latest --mcp your-mcp --dry-run
# Test template installation
npx claude-code-templates@latest --template your-template --dry-run
# Test with specific scenarios
npm start -- --language python --framework django --dry-run
npm start -- --language javascript --framework react --dry-run
# Test analytics
npm run analytics:test
# Test chat monitor
npm run chats:test
# Test health check
npm run health-check:test
git clone https://github.com/your-username/claude-code-templates.git
cd claude-code-templates
git checkout -b feature/your-contribution
cd cli-tool
npm test
npm start -- --dry-run
By contributing to this project, you agree that your contributions will be licensed under the MIT License.
All contributors are recognized in our:
Thank you for helping make Claude Code Templates better for everyone! 🚀
Then run bash setup.sh. It stops at the first error and is safe to run again. Servers and tests aren't started; it lists them at the end. Read it first.
git clone https://github.com/davila7/claude-code-templates.git cd claude-code-templates
Node.js 22
# needs nvm: https://github.com/nvm-sh/nvm nvm install 22
Fill in secrets: PUBLIC_CLERK_PUBLISHABLE_KEY, CLERK_SECRET_KEY, GITHUB_CLIENT_SECRET, SUPABASE_SERVICE_ROLE_KEY, RAPIDAPI_KEY, SERPAPI_KEY, GOOGLE_API_KEY, SUPABASE_API_KEY
cp .env.example .env
npm ci
npm test
Runs with npx, no global install needed.
# Browse the catalog and install interactively
npx claude-code-templates@latest
# Install one component
npx claude-code-templates@latest --skill creative-design/frontend-design
# Install a stack in one command
npx claude-code-templates@latest \
--agent development-team/frontend-developer \
--command testing/generate-tests \
--mcp integration/github-integration
Add --yes to skip the confirmation prompt. Every component page on aitmpl.com shows its exact install command. To find one from the terminal: npx claude-code-templates@latest --search "code review".
npx claude-code-templates@latest --skill creative-design/frontend-design
npx claude-code-templates@latest --agent development-tools/code-reviewer
npx claude-code-templates@latest --command performance/optimize-bundle-size
npx claude-code-templates@latest --mcp database/postgresql-integration
npx claude-code-templates@latest --setting mcp/mcp-timeouts
npx claude-code-templates@latest --hook git/conventional-commits
npx claude-code-templates@latest --mod security/secret-redactor # Claude Code 2.1.287+
npx claude-code-templates@latest --loop engineering/docs-sweep-loop # also installs the components the loop uses
npx claude-code-templates@latest --plugin productivity/champion-kit # via the aitmpl marketplace, Claude Code 2.1.288+
</details>No reports for Linux yet.
GitManual AI Agent reads the 12 key files and writes a step-by-step guide for Linux. Once written, it's saved for everyone.